This application is for Dr. Sanjay Basu, a physician-epidemiologist establishing himself in the field of statistical and modeling research on cardiovascular disease disparities. This award will allow Dr. Basu to: (1) develop a career focus in nutritional epidemiology and cardiovascular disease disparities; (2) learn how to implement systems science mathematical modeling methods that simulate the impact of cardiovascular disease interventions; and (3) gain the necessary content expertise to address major unanswered questions in the epidemiology of cardiovascular disease disparities. To achieve these goals, Dr. Basu has assembled a mentoring team with expertise on nutrition and its relationship to cardiovascular disease risk; the epidemiology of cardiovascular disease disparities; and mathematical modeling. Dr. Basu's research proposal addresses disparities in the prevalence of hypertension between groups of different socioeconomic status (SES). The Institute of Medicine (IOM) has declared lower sodium consumption to be a national goal to reduce hypertension-related cardiovascular disease.2 Dr. Basu's research will test hypotheses about the relationship between social factors and sodium consumption among low-SES populations experiencing the highest prevalence of uncontrolled hypertension.3 In Aim 1, a regression tree analysis4 applied to the National Health and Nutrition Examination Survey will test the hypothesis that variations in sodium consumption within the population can be predicted using a few key sociodemographic variables.5 The analysis will identify what social factors can best explain heterogeneities in sodium intake within the US population. In Aim 2, we will then test the hypothesis that an economic program incentivizing low-SES populations to increase the number of times per month they go grocery shopping can lead high-sodium consumers to substitute packaged products with fresh produce, thereby lowering their sodium consumption.6-8 To better assess causality, we will analyze a unique quasi-natural experiment in Delaware, New Jersey and Pennsylvania in which a subset of households on food stamps were randomly allocated to receive benefits twice monthly instead of once monthly, which induced more frequent grocery shopping despite having the same overall monthly grocery budget as matched controls. The results of Aims 1 and 2 will be incorporated into a simulation model of hypertension in Aim 3. An agent-based model will be constructed to compare the impact of the program in Aim 2 to two other programs- front-of-package nutrition labeling9 and a vegetable voucher plan10-that attempt to influence consumers to reduce sodium consumption. The model will estimate the impact and cost-effectiveness of the interventions for reducing national hypertension disparities. This will form the basis of an R01 application to study novel sodium reduction strategies that target consumer food choice behaviors. This proposal addresses NHLBI strategic plan challenge 3.1, which encourages social science research incorporating systems science modeling methods to investigate social factors and policy interventions to reduce health disparities.11